发表机构
Carnegie Mellon University; Rice University（卡内基梅隆大学; 莱斯大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对长视频生成中历史上下文难以压缩的问题，提出预测对齐上下文压缩（PACC），利用冻结生成器通过在线策略蒸馏训练压缩器，在MBench和VBench-Long上显著提升记忆性能且不修改生成器。
AI 中文摘要
标准视频生成器并不能原生地将历史上下文压缩为可复用的记忆令牌。随着生成的持续进行，不断增长的历史使得由于长上下文退化，从早期帧中保留信息变得越来越困难。基于关键帧的方法通过保留选定的过去帧来应对这一挑战，但可能会丢弃未来生成所需的信息。我们研究的是，与其仅依赖帧选择，一个冻结的视频生成器能否提供学习历史紧凑表示所需的监督。我们提出了预测对齐上下文压缩（Prediction-Aligned Context Compaction, PACC），该方法使用一个学习到的压缩器将过去帧中的信息聚合为紧凑的记忆令牌。我们通过在线策略蒸馏来训练压缩器，在压缩记忆条件下使用同一个冻结生成器作为学生，而在完整历史条件下作为教师。学生生成后续内容，而教师在每个去噪步骤中为相同的噪声输入提供目标。仅更新压缩器，以使学生的预测与这些目标对齐。我们在MBench上评估了PACC，该基准联合衡量记忆事件覆盖率和一致性。PACC在Causal-rCM上超过最强基线6.63个百分点，在Causal Forcing上超过3.19个百分点。使用MovieGen提示在VBench-Long上的评估进一步表明，PACC生成的分钟级视频在生成质量上与基线相当。这些结果共同表明，学习压缩历史上下文可以在不修改底层生成器的情况下改善长视频记忆。
英文摘要
Standard video generators do not natively compact historical context into reusable memory tokens. As generation continues, the growing history makes it increasingly difficult to retain information from earlier frames due to long-context degradation. Key-frame-based approaches address this challenge by retaining selected past frames, but can discard information needed for future generation. Rather than relying on frame selection alone, we study whether a frozen video generator can supply the supervision needed to learn a compact representation of the history. We propose Prediction-Aligned Context Compaction (PACC), which uses a learned compressor to aggregate information across past frames into compact memory tokens. We train the compressor through on-policy distillation, using the same frozen generator both as a student when conditioned on compressed memory and as a teacher when conditioned on the full history. The student generates continuations, while the teacher provides targets for the same noisy inputs at each denoising step. Only the compressor is updated to align the student's predictions with these targets. We evaluate PACC on MBench, which jointly measures memory-event coverage and consistency. PACC outperforms the strongest baseline by 6.63 points on Causal-rCM and 3.19 points on Causal Forcing. Evaluation on VBench-Long using MovieGen prompts further shows that PACC produces minute-long videos with generation quality competitive with baselines. Together, these results show that learning to compact historical context can improve long-video memory without modifying the underlying generator.
CommentsUnder Review
